##### Scenario Building a real-time fraud-detection system for subscription payments. ##### Question Outline the end-to-end workflow for training and deploying an XGBoost model to flag fraudulent transactions in real time. Which evaluation metrics would you prioritize and why? How would you handle severe class imbalance during training? Describe one strategy for monitoring model drift after deployment. ##### Hints Cover feature engineering, class weighting or sampling, precision-recall trade-offs, and online monitoring.

This interview question evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer for Design Real-Time Fraud Detection with XGBoost Model states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

How do I approach Machine Learning interview questions?

Machine Learning questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master machine learning interviews.

What difficulty level is this interview question?

This is a medium difficulty Machine Learning question, commonly asked during Other rounds at Netflix.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates at Netflix during technical interviews.

Design Real-Time Fraud Detection with XGBoost Model

Real-Time Fraud Detection with XGBoost (Subscription Payments)

Scenario

You need to build and operate a real-time system that flags potentially fraudulent subscription-payment transactions with sub-second latency. Historical labels come from chargebacks/refunds with a delay of weeks. Data includes transaction attributes, user/account metadata, device/network signals, and historical behavior.

Task

Outline the end-to-end approach, covering:

End-to-end workflow

Data ingestion, labeling, feature engineering (batch + streaming), training/validation protocol, hyperparameter tuning, offline–online feature parity, deployment architecture, and a feedback loop.

Evaluation metrics

Which metrics you would prioritize in an imbalanced, high-stakes setting and why.

Handling severe class imbalance

Approaches such as class weighting, sampling, threshold tuning, and any loss/metric choices.

Monitoring for model drift post-deployment

Describe one concrete strategy to detect and respond to drift.

Constraints & Assumptions

Preserve the scope, facts, inputs, and requested outputs from the prompt above.
If the prompt leaves a detail unspecified, state a reasonable assumption before relying on it.
Keep the answer interview-ready: concise enough to present, but concrete enough to implement or evaluate.

Clarifying Questions to Ask Guidance

Clarify the task, data shape, labels, constraints, and evaluation metric.
State assumptions behind the math or modeling technique you choose.
Connect theory to practical training, debugging, and deployment implications.

What a Strong Answer Covers Guidance

Correct definitions and formulas where the prompt requires them.
A practical explanation of how the method behaves on real data.
Trade-offs, failure modes, diagnostics, and mitigation strategies.
Evaluation choices that match the product or modeling objective.

Follow-up Questions Guidance

How would noisy labels, class imbalance, or distribution shift affect the answer?
What would you monitor after deployment?
Which baseline would you compare against first?